Open Access
Add to BibSonomy
Abstract
Amorphous Ge-As-Se thin films have been prepared by the magnetron sputtering deposition technique, and the photo-induced effects (PIEs) in the different states have been investigated. It was found that, for the as-deposited films, Ge5 exhibits photodarkening (PD) while Ge15 and Ge25 undergo photobleaching (PB), and the degree of PB in Ge25 is larger than that in Ge15. On the other hand, all the annealing films exhibit PD, and the degree of PD decreases from Ge5 to Ge25. In all cases, PD is reversible while PB is irreversible. The Ge/As ratio or the lone pair electrons in Se atoms that were suggested for PIEs in the chalcogenide films cannot account for the present results in the GeAsSe films. Nevertheless, Ge15 exhibits minimum PIEs during a continuous illumination process that could be the best option for waveguide fabrication.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Due to amorphous nature of chalcogenide glasses, structure relaxation usually occurs with a prolonged storage time [1,2]. Moreover, structural relaxation can be accelerated via thermal annealing or optical and ion irradiation [3,4]. This is usually even more pronounced in thin films which, compared with bulk glasses, have been formed in nonequilibrium conditions as an atomic vapor or cluster condenses onto a cold substrate. More wrong bonds in the films are often formed which is different from those created in a bulk glass produced in near equilibrium conditions by a conventional melt-quenching method [3,4]. However, optical devices like waveguides are usually based on high quality thin films [5,6], therefore it is critically to investigate the various thermal- and photo-induced effects (PIEs) in the films.
Regarding PIEs, Tanaka et al. [1] have emphasized the difference in as-prepared and annealed films in terms of quasi-stability and structural disorder. The annealed film has the lowest free energy and smallest disorder, and photo-excitation always produces more unstable and more disordered states, therefore PIE is reversible. On the other hand, the as-prepared films are likely to be energetically higher or lower than the illuminated state, therefore, illumination effects become irreversible [1]. Irreversible or Reversible change are defined as whether PIEs can be erased by thermal annealing at some temperatures.
Among various PIEs, the most notable ones are photodarkening (PD) and photobleaching (PB), which correspond to the decrease or the increase of the optical band gap energy via light shining, respectively [7,8]. PIEs have been widely used in the applications such as optical writing, photolithography, etc. [1,2]. While PIEs are beneficial to these applications, obviously it is undesirable for other applications like optical micro-lens for infrared optics. Therefore, it is potentially important to understand how to tune PIEs via the design of material compositions in order to meet the different requirements for various applications.
Our on-going efforts have concentrated on exploring high quality thin films used in chalcogenide-based optical waveguide. Recent investigations on more than 30 pieces of Ge-As-Se bulk glasses suggested that, the glasses with a mean coordination number (MCN-the sum of the products of the valence times the abundance of the individual atomic constitutes) around 2.45-2.5 had a minimum glass transition activation energy and fragility index, therefore the glasses were expected to have minimum driving force for structural relaxation [9]. Although the results are associated with the bulk glasses, it provides a guideline to screen the best materials for the applications in photonics. For Ge-As-Se thin films prepared by thermal evaporation, in the early 1970s, Igo et al. reported the existence of two effects: one is thermal and the other is optical. The thermal effect always shifts the absorption edge to shorter wavelengths which is later defined as PB, and the optical effect shifts it to longer wavelengths which is defined as PD. There is a competition between PD and PB depending on Ge contents in GeAsSe films [10–12]. This opens up the possibility to tune PIEs in GeAsSe films for various applications. Following that, recently Yang et al. [13] reported a photo-stable film with a composition of Ge10As35Se55, while Su et al. [3] showed that the films with a mean coordination number around 2.5 were most stable. Barik et al. [14] observed the kinetics of photodarkening in GexAs45-xSe55 glasses when the network rigidity was increased with x from 0 to 16. Kumar et al. [15] observed the transition from PD to PB when the composition of the film changed from Ge-poor to Ge-rich in GexSe100-x thin films. While all the films used in [3,4,13–15] were deposited by thermal evaporation, Nemec et al. also reported photosensitivity in pulsed laser deposited GeAsSe films and found that Ge20As20Se60 films exhibited better photostability in terms of almost zero photorefraction in relaxed states [8,16].
Therefore, in this paper, we prepared GeAsSe films with three different compositions having an MCN that is less, close and more than 2.5. Based on the rigidity theory, three compositions span floppy, intermediate and rigid phases in the chalcogenide glasses [1,2]. We fixed As concentration at 24% and then change Ge and Se concentrations in the films, therefore it is much easier to probe the effect of Ge on PIEs in the GeAsSe films. Furthermore, we employed magnetron sputtering to prepare the films since sputtering can produce much dense films while evaporation produces more disordered/fragmental/molecular films with voids and wrong bonds. Therefore, we could see whether the films produced by the different deposition methods could exhibit different PIEs. We investigated the transmission of the film before and after illumination and annealing in order to understand how the PIEs can change in the films with different MCN. The kinetics process of PIEs was modelling with a time dependent exponential function, and the possible mechanism was analyzed.
2. Experiments
We prepared high purity Ge-As-Se bulk glasses with different compositions, e.g., Ge3As24Se73, Ge11.5As24Se64.5 and Ge20As24Se56, and used them as sputtering targets. Thin films were deposited onto microscope glass and Si substrates by the magnetron sputtering deposition technique. The thicknesses of these films were in-situ controlled by a thickness monitor equipped in the chamber and further checked by Veeco Dektak 150 surface profiler. The final values of the thickness are listed in the Table 1. The chemical compositions of the films were determined by an Energy Dispersive X-ray Spectrometer (EDS) installed in a Tescan VEGA3 SB-Easyprobe Scanning Electron Microscope (SEM). For convenience, individual chemical compositions of these films were also shown in Table 1.
Table 1. The thicknesses and compositions of these films and their corresponding value of MCN
We used the pump-probe method to study the photo-induced effects in these three films. The experimental set up is same as those in previous studies [17]. In brief, we chose high power collimated light emitting diode with a center wavelength of 655 nm (1.89 eV) and bandwidth extending from 625 nm to 690 nm (1.98 eV–1.79 eV) at the 10% intensity level as an illumination beam with a beam diameter of ∼5 mm. The illumination power was tuned by the filter in front of the films. The probe beam was a low intensity white light with a beam diameter of 2 mm and a wavelength regime of 550–1000 nm. The change in transmission before and after the pump beam illumination was recorded in an interval of 10 ms using a high-resolution optical absorption spectrometer (Ocean Optics HR2000).
The structure of the films was analyzed using micro-Raman spectrometer. Raman spectra were recorded at room temperature by a 785 nm laser excitation with an InVia spectroscope (Renishaw) coupled to a Leica DM 2500 M microscope using a x50 magnification objective.
3. Results and discussion
To distinguish these different PIEs, the as-deposited films were illuminated or annealed, and then the annealed films were illuminated again, following by the re-annealing. For thermal annealing, we measured glass transition temperature Tg of both the bulk glasses and the films using a flash DSC where only tens of nanograms of materials are needed. Therefore, it is possible to just scratch piece of the films for Tg measurement. We found the difference of Tg between the bulk and film is less than 5°C. Therefore, we annealed the films at a temperature that is 20°C below their respective Tg. The annealing temperature is 130°C, 210°C, and 310°C for Ge5, Ge15, and Ge25, respectively. Thermal annealing was performed in a vacuum oven with a pressure of 10−3 Torr. We raised the temperature at a rate of 1°C/min to the setting temperature, hold it for 5 hours, and finally reduced the temperature to room temperature at a rate of 1°C/min. All the illumination was performed in air.
Transmission spectra of the films under these five different states were measured, and the results were shown in Figs. 1(a), 1(b) and 1(c), for Ge5, Ge15 and Ge25 films, respectively.
Fig. 1. Transmission spectra of the as-prepared and illuminated Ge-As-Se thin film (a) Ge5 (b) Ge15 (c) Ge25.
It is clearly seen that, compared with those in the as-prepared films, the transmission edge of the illuminated film shows a redshift (e.g., shift to longer wavelengths) in Ge5, no change in Ge15 and a blueshift (e.g., shift to shorter wavelengths) in Ge25 film, respectively. Annealing of the films results in no change of the transmission edge in Ge5, blueshift in Ge15 and Ge25, respectively. On the other hand, while the annealing films are used as the initial state for further illumination, the transmission edge of all the illuminated film show a redshift. Moreover, re-annealing of these annealing-illumination films can recover the transmission edge to their origin states (annealing films), clearly demonstrating a reversible PD effect in GeAsSe films.
We used the Tauc plots to determine the optical bandgap of the films under various states, and the results were shown in Fig. 2, and the values of the optical bandgap under various states were also listed in Table 2. From the change of the optical bandgap energy, we can conclude that for Ge5 film, illumination of the as-prepared films leads to a decrease of the bandgap energy from 1.85 eV to 1.83 eV, and thus this is a PD process which is reversible as evident by the annealing/illumination/re-annealing cycle. However, Ge15 and Ge25 exhibit different behaviors. Illumination of the as-prepared films leads to an increase of the bandgap energy, and thus being a PB process in both cases which is irreversible since both bandgap energy cannot be recovered to their respective as-prepared state. Moreover, illumination of both the annealed Ge15 and Ge25 exhibit a decrease in the optical bandgap energy, and thus being a PD process, which is reversible since both bandgap energy can be recovered to their respective annealing state during an annealing/illumination/re-annealing cycle. Nevertheless, Ge15 and Ge25 with high Ge content show a larger change of the optical bandgap in PB and PD processes in the different states. This is in agreement with the previous literatures [11,15].
Fig. 2. Tauc plot of (αħν)1/2 vs. ħν for (a) Ge5, (b)Ge15, and (c)Ge25, respectively. Each inset corresponds to the enlarged shade area in the respective main panel showing the optical band gap of the films under various states.
Table 2. The bandgap energy of the films under various states extracted from the Tauc plots
It is essential to avoid any thermal effect during the investigation of PIEs. Schardt et al. [18] reported that, a temperature rise in the films illuminated by 800 nm continuous laser with an energy of 0.5-40 W/cm2 is less than 5 K, and our previous results in the films illuminated by 655 nm laser with a power of 1 W/cm2 also showed the temperature raise is less than 3 K detected by infrared thermal camera images [17]. Therefore, thermal diffusion induced by illumination is not expected to take place in our experiments. This is also one of the reasons why the illumination of the films was performed in air without any protection by the inert gases of N2 or Ar gas flow in our experiments.
We investigated the kinetics of PIEs in the as-prepared films, and recorded the transmission spectrum of the as-prepared film in dark condition and denoted it as Ti. Next, we turned on the pump beam and simultaneously recorded the transmission spectrum as a function of time (Tf). The data showed temporal evolution of transmission ratio (Tf/Ti) for three films at probe wavelengths for which transmission was 20% of the value in the dark and as-prepared condition. The results are shown in Fig. 3, where all Tf/Ti decrease in the initial illumination process within 100s. However, for Ge5 film, Tf/Ti decreases to 0.7 and then almost constant with prolonged illumination time up to 12000 s. For Ge15 film, Tf/Ti almost recovers to 1 after initial decrease, while Ge25 film exhibits obvious PB in the prolonged illumination.
Fig. 3. Time evolution of Tf /Ti for Ge5, Ge15, and Ge25 films illuminated by a 655 nm laser with a power intensity of 100 mW/cm2.
Previously, Khan et al. [19] reported a crossover from PD to PB in GexAs35-xSe65 films and a coexistence of PD and PB in Ge19As21Se60 films. They claimed that the role of Ge:As ratio in controlling PD and PB behaviors where the films exhibit PD at x < 15% while PB at x > 15% in GexAs65-2xSe65. Mishchenko et al. [20] also investigated the dynamic variations of PIEs in GexSe100-x films. It was found that, there is a critical concentration of Ge around 30% that corresponds to the crossover from transient PD to the mixture of transient PD and metastable PB. Our results in the unannealed films in Fig. 3 are in qualitative agreement with those in the literatures [19,20].
We also measured the kinetics effect of PIEs in the annealed films as shown in Fig. 4. Three films show same PD behavior, although the time required to relax themselves to their respectively saturated states is different. We fitted the whole process using a stretched exponential function that describes PD [19]:
where the subscript d corresponds to PD. ΔTsd, ${\tau _d}\left\langle {it} \right\rangle ,\, \left\langle {/it} \right\rangle {\beta _d}$, t and A are the metastable part, the time constant, the dimensionless parameter, the illumination time, and a temperature dependent quantity that is equal to the maximum transient changes, respectively. The value between the initial and final process of Tf/Ti, which was determined by the fitting stretched exponential function, was defined as ΔT. The fitting parameters based on Eq. (1) are listed in Table 3. Since the illumination time is not long enough, the parameters of the fits of our experimental data for Ge25 film might not be insufficiently accurate for a quantitative analysis.Fig. 4. The time dependence of Tf/Ti in annealed Ge5, Ge15 and Ge25 films. The black lines are the experimental results while the red ones are their respective fitting curves based on Eq. (1).
Table 3. The fitting parameters for PIEs in annealed Ge5, Ge15 and Ge25 films
We noted the fact that the degree of PD is sensitive to the illumination intensity and photon energy. Here, illumination photo energy of 1.89 eV (655 nm) is close or less than the bandgap energy of the films in the different states as shown in Table 1, and thus the results observed are induced by bandgap light. We further investigated the sensitivity of PIEs to the different illumination power, and the typical results for the as-deposited and annealed Ge15 films are shown in Figs. 5(a) and 5(b), respectively. In Fig. 5(a) for the as-deposited film, while Tf/Ti exhibits a transient PD in the initial illumination, it recovers to almost 1 in all cases. Moreover, switching-off the light can induce an increase of Tf/Ti and such an increase becomes larger with increasing illumination power. Re-switching-on light can recover Tf/Ti to its initial value. Such switching-on/off behaviours are same as those in [19–22]. For the annealed films in Fig. 5(b), it was found that, in all cases, the PD behaviours can be saturated with few thousand seconds illumination, the degree of the decay of Tf/Ti is increased from 0.15 to 0.4 with increasing illumination power from 10 to 400 mW/cm2. Nevertheless, switching- on/off cycles of the illumination can also recover Tf/Ti to its initial value as evident by the curve with 100 mW/cm2 illumination in Fig. 5(b).
Fig. 5. Photo-induced effects in as-deposited (a) and annealed (b) Ge15 films under illumination with different power.
The present results provide clear evidence that, while the as-prepared films exhibit so-called transient PD and PB, we cannot find any PB in the annealed films, instead, only PD can be observed in the annealed films although the degree of the PD is different in the films with different composition. Moreover, all the PD processes are reversible while all the PB processes are irreversible, as evident by the data in Table 1. Regarding the origin of these PIEs, it has been proposed that, PD arises from photo-induced broadening in the top of the valence-band, and PB arises from homo-to-hetero polar bond switching and oxidation of As and Ge, and the switching of the bonds becomes greater in As- and/or Ge-rich films [1]. We also note two interesting results that were reported recently. Pritam et al. [19] emphasized the role Ge:As ratio in controlling PIES in GeAsSe films. From Raman spectra, they claimed that, instantaneous PD arises from the As part of the network, and PB from the Ge part of the network. Mishchenko et al. [20] investigated PIEs in GeSe films. Combining Raman analysis with first-principles simulations, they concluded that, alteration in PIEs is governed by availability of the lone pair states at the Se atoms that are sensitive to the Ge:Se ratio. However, it is still arguable, since the structural analysis from Raman scattering spectra requires the deconvolution of the spectra into the different structural units. While some structural units with large weight factor (depending on the composition and vibrational intensity) could be clearly presented in the spectra, some of the vibrations with subtle amounts could be missing, but these could play important role in determining PIEs. Therefore, Raman results are not sufficient to gain an understanding about the origin of the transient PD and changes in the topology.
We also recorded Raman spectra of the films before and after irradiation/annealing, but no obvious structural change can be observed. Moreover, Ge(As)-O vibrations cannot be observed in any states of the films using Infrared and Raman spectra. One of the physical origins of PB is ascribed to the existence of the oxide states of Ge/As [23], but the present results observed in the films prepared by magnetic sputtering seem to rule out such a possibility. We also note a fact that, in [13,18–21], the films were illuminated by the above-bandgap light which is in contrast with that by bandgap light in our case, and thus the effect of illumination light with different photo energy on PIEs could be different and the mechanism for PIEs proposed in [8,9,17] could not be simply applied to our cases. While we observed the similar PIEs effects in the as-prepared films as those in [13,18–21], the key results in the paper are that, the annealed Ge-As-Se films only exhibit PD, not sensitive to the any Ge:As ratio or any changes in the topology since the film compositions in our paper span from the floppy to rigid phase. On the other hand, we found that, there is no change of the films compositions detected by EDS before and after illumination/annealing, and thus the change of the number of the lone pair states should be negligible although the illumination or annealing could annihilate some of the dangling bonds in the as-deposited films. We note that in Table 2, the bandgap decreases 0.3 eV for both Ge5 and Ge15 but 0.7 eV for Ge25 from their respective annealed to annealed-illuminated state. Therefore, it appears that PD is not determined only by the lone pair states at the Se atoms. Further investigation is required and can be done by fine structural analysis methods to establish the correlation between the light induced effect and film structure.
Nevertheless, the present results also provide some clues for the choice of the glasses for waveguide fabrication. When photoresists are covered in the films during the fabrication, annealing procedure is usually omitted since this could induce additional issues. That is the reason why PD rather than PB in the fresh films could be of certain interest because of the opportunity of choosing the appropriate solvent [24,25]. In terms of this, it seems that, Ge15 films exhibit minimum PIEs during a continuous illumination process compared with Ge5 and Ge25 as shown in Fig. 2. Therefore, such composition could be appropriate for the waveguide fabrication. This is in excellent agreement with the previous results, where the bulk glasses with an MCN around 2.45-2.5 had a minimum fragility index, and thus minimum driving force for structural relaxation [9].
4. Conclusion
We prepared three Ge-As-Se films with different compositions crossing the floppy, intermediate and rigid phases in chalcogenide glasses and investigated PIEs in the films with various states. We found that, the as-deposited films exhibit similar PIEs in [13,18–21], e.g., Ge5 undergoes reversible redshift (PD) while Ge15 and Ge25 exhibit irreversible blueshift (PB), and the degree of PB in Ge 25 is larger than that in Ge15. On the other hand, all the annealing films exhibit PD, and the degree of PD decreases from Ge5 to Ge25. While the underlying mechanism is unclear, it appears that, Ge/As ratio or the lone pair electrons in Se atoms that was suggested for the photoinduced effect in chalcogenide film cannot account for the present results. Nevertheless, during a continuous illumination process, Ge15 exhibits minimum PIEs, and thus is the best choice of the glass composition for waveguide fabrication.
Funding
National Natural Science Foundation of China (Nos. 61775109, Nos. 11847159); Natural Science Foundation of Henan Province (Nos.2019JJ50410); Education Department of Hainan Province (Nos.18C0744).
References
1. K. Tanaka and K. Shimakawa, Amorphous Chalcogenide Semiconductors and Related Materials (Springer, 2011).
2. R. P. Wang, Amorphous Chalcogenides: Advances and Applications (Pan Stanford Publishing, 2014).
3. X. Su, R. Wang, B. Lutherdavies, and L. Wang, “The dependence of photosensitivity on composition for thin films of GexAsySe1-x-y chalcogenide glasses,” Appl. Phys. A 113(3), 575–581 (2013). [CrossRef]
4. D. A. P. Bulla, R. P. Wang, A. Prasad, A. V. Rode, S. J. Madden, and B. Luther-Davies, “On the properties and stability of thermally evaporated Ge-As-Se thin films,” Appl. Phys. A 96(3), 615–625 (2009). [CrossRef]
5. S. J. Madden, D. Y. Choi, D. A. Bulla, A. V. Rode, and B. J. Eggleton, “Long, low loss etched As2S3 chalcogenide waveguides for all-optical signal regeneration,” Opt. Express 15(22), 14414–14421 (2007). [CrossRef]
6. Yi Yu, Xin Gai, Pan Ma, Khu Vu, Zhiyong Yang, Rongping Wang, Duk-Yong Choi, Steve Madden, and Barry Luther-Davies, “Experimental Demonstration of Linearly Polarized 2-10 µm Supercontinuum Generation in a Chalcogenide Rib Waveguide,” Opt. Lett. 41(5), 958–961 (2016). [CrossRef]
7. K. Shimakawa, A. Kolobov, and S. R. Elliott, “Photoinduced effects and metastability in amorphous semiconductors and insulators,” Adv. Phys. 44(6), 475–588 (1995). [CrossRef]
8. P. Hawlova, M. Olivier, F. Verger, V. Nazabal, and P. Nemec, “Photosensitivity of pulsed laser deposited Ge20As20Se60 and Ge10As30Se60 amorphous thin films,” Mater. Res. Bull. 48(10), 3860–3864 (2013). [CrossRef]
9. T. Wang, O. Gulbiten, R. Wang, Z. Yang, A. Smith, B. Luther-Davies, and P. Lucas, “Relative contribution of stoichiometry and mean coordination towards the fragility of Ge-As-Se glass forming liquids,” J. Phys. Chem. B 118(5), 1436–1442 (2014). [CrossRef]
10. T. Igo and Y. Toyoshima, “Optically induced reversible change in amorphous semiconductors,” Jpn. J. Appl. Phys. 11(1), 117–118 (1972). [CrossRef]
11. T. Igo and Y. Toyoshima, “A reversible optical change in the As-Se-Ge glasses,” J. Non-Cryst. Solids 11(4), 304–308 (1973). [CrossRef]
12. K. Tanaka and K. Shimakawa, “Chalcogenide glass in Japan: A review on photoinduced phenomena,” Phys. Status Solidi B 246(8), 1744–1757 (2009). [CrossRef]
13. G. Yang, H. Jain, A. Ganjoo, D. Zhao, Y. Xu, H. Zeng, and G. Chen, “A photo-stable chalcogenide glass,” Opt. Express 16(14), 10565 (2008). [CrossRef]
14. A. R. Barik, K. V. Adarsh, R. Naik, R. Ganesan, and K. Shimakawa, “Role of rigidity and temperature in the kinetics of photodarkening in GexAs45-xSe55 thin films,” Opt. Express 19(14), 13158–13163 (2011). [CrossRef]
15. R. R. Kumar, A. R. Barik, E. M. Vinod, M. Bapna, K. S. Sangunni, and K. V. Adarsh, “Crossover from photodarkening to photobleaching in a-GexSe100-x thin films,” Opt. Lett. 38(10), 1682–1684 (2013). [CrossRef]
16. P. Němec, S. Zhang, V. Nazabal, K. Fedus, and X. H. Zhang, “Photo-stability of pulsed laser deposited GexAsySe100-x-y amorphous thin films,” Opt. Express 18(22), 22944–22957 (2010). [CrossRef]
17. S. Zhang, Y. Chen, R. Wang, X. Shen, and S. Dai, “Observation of photobleaching in Ge-deficient Ge16.8Se83.2 chalcogenide thin film with prolonged irradiation,” Sci. Rep. 7(1), 14585 (2017). [CrossRef]
18. C. R. Schardt, P. Lucas, A. Doraiswamy, P. Jivaganont, and J. H. Simmons, “Raman temperature measurement during photostructural changes in GexSe1−x glass,” J. Non-Cryst. Solids 351(19-20), 1653–1657 (2005). [CrossRef]
19. P. Khan, H. Jain, and K. V. Adarsh, “Role of Ge:As ratio in controlling the light-induced response of a-GexAs35-2xSe65 thin films,” Sci. Rep. 4(1), 4029 (2015). [CrossRef]
20. A. Mishchenko, J. Berashevich, K. Wolf, D. A. Tenne, A. Reznik, and M. Mitkova, “Dynamic variations of the light-induced effects in a-GexSe100-x films: experiment and simulation,” Opt. Mater. Express 5(2), 295–306 (2015). [CrossRef]
21. P. Khan, A. R. Barik, E. M. Vinod, K. S. Sangunni, H. Jain, and K. V. Adarsh, “Coexistence of fast photodarkening and slow photobleaching in Ge19As21Se60 thin films,” Opt. Express 20(11), 12416–12421 (2012). [CrossRef]
22. V. Lyubin, M. Klebanov, A. Burner, N. Shitrit, and B. Sfez, “Transient photodarkening and photobleaching in glassy GeSe2 films,” Opt. Mater. 33(6), 949–952 (2011). [CrossRef]
23. Q Yan, H. Jain, J. Ren, D. Zhao, and G. Chen, “Effect of photo-oxidation on plotobleaching of GeSe2 ad Ge2Se3 thin films,” J. Phys. Chem. C 115(43), 21390–21395 (2011). [CrossRef]
24. E. Skordeva, D. Arsova, Z. Aneva, N. Vuchkov, and D. Astadjov, “Laser induced photodarkening and photobleaching in Ge-As-S thin films,” Proc. SPIE 4397, 348–352 (2001). [CrossRef]
25. D. A. Turnbull, J. S. Sanghera, V. Q. Nguyen, and I. D. Aggarwal, “Fabrication of waveguides in sputtered films of GeAsSe glass via photodarkening with above bandgap light,” Mater. Lett. 58(1-2), 51–54 (2004). [CrossRef]
